Treating HIV: When to Start, Who Should Treat, and Still Missing Many

June 1, 2008

Apart from clinical trials that provide highly active antiretroviral therapy shortly after the recognition of an acute seroconversion reaction-an intervention that some believe could be a prelude to eradicating HIV -there has long been debate over the optimal time to initiate antiretroviral therapy.

Apart from clinical trials that provide highly active antiretroviral therapy shortly after the recognition of an acute seroconversion reaction-an intervention that some believe could be a prelude to eradicating HIV 1 -there has long been debate over the optimal time to initiate antiretroviral therapy. There are cogent arguments for both early and late starts, given the difficulty in maintaining the near-perfect adherence that is required by many regimens and the concern over cardiovascular, bone, and other metabolic adverse effects of the drugs used versus the risk of leaving patients vulnerable to irreversible immune compromise and premature death if therapy is inappropriately withheld. Clinical trials are unlikely to provide a definitive answer anytime soon, but a recent computer simulation, based on the Veterans Aging Cohort, questions the premises that favor delayed treatment in some situations, particularly for persons 30 to 39 years of age.2

As noted by a recent NIH-convened panel,3 these arguments persist because it is difficult to quantitate the risks and benefits of early treatment-an accurate assessment requires many years of follow-up. The newest Veterans’ simulation, involving 5742 HIV-infected patients and 11,484 matched uninfected controls seen from 1997 through 2004, considered factors that were omitted from earlier mathematical models. These included wider age ranges (30 years to more than 60 years), site of care and comorbid conditions, accrual of resistance mutations, drug toxicities and adverse effects, and preference-based quality-of-life measures.

The HIV-positive cohort was selected for an initial low risk of HIV-related death based on a threshold CD4+ T-cell count of 500/µL. Limitations to the generalization of results include the male predominance (98%) and the fact that different types of highly active antiretroviral therapy-particularly protease inhibitor (PI)- versus NNRTI-based regimens-were not factored into the analysis. (PI-based regimens are linked to an increased risk of premature atherosclerosis and related cardiovascular disease compared with NNRTI-based therapy.4)

Overall, highly active antiretroviral therapy was associated with a 3.8-fold (range, 3.1- to 4.6-fold) increased hazard of non-HIV-related mortality.2 This included an adjusted hazard ratio of 2.3 for those 40 to 49 years of age; it was more than double that risk, at 5.4, for those 60 years and older. Non-AIDS cancer was the highest-risk comorbid condition, with a 2.4-fold risk, followed by congestive heart failure at 2.2-fold.

Two-thirds of patients reported adverse effects that they “possibly or definitely” attributed to highly active antiretroviral therapy. The overall magnitude of adverse effects in these patients, recorded as a utility score of 0.08 on a scale of 0 to 1.0, was comparable to that in patients with mild cardiac angina.2

In terms of mortality, the simulation model used had been validated in prior clinical cohorts and accurately predicted time to treatment failure and survival.5 It now showed that earlier therapy-starting at CD4+ cell counts with an upper limit of 500/µL versus 350/µL or 200/µL-improved life expectancy in many of the scenarios evaluated, despite the fact that it hastened accumulation of resistance mutations and reduced future drug options. For 30-year-olds, earlier initiation was always favored, regardless of HIV RNA level (10,000 to 300,000 copies/mL or higher). For 50-year-olds, however, harm from early initiation outweighed the benefits in most instances.

Computer simulations favored withholding treatment in asymptomatic patients 50 years and older unless the CD4+ cell count was 200/µL or less, or the HIV RNA level was 300,000 copies/mL or higher and CD4+ cell count was 350/µL or less. Results for those aged 40 to 49 years were more varied and generally favored early treatment only for those with higher HIV RNA levels. The quality-adjusted life expectancy results paralleled the life expectancy results in this simulation.

The authors argued that these data should inform clinical care, adding evidence to help clarify the more nebulous sections of current treatment guidelines. For example, those guidelines cryptically suggest that for patients with CD4+ cell counts of more than 350/µL and less than 500/µL and HIV RNA levels of 100,000 copies/mL or higher, one should "Defer [HAART], but treating may be appropriate."3 The computer model suggests deferring treatment only if the patient is 50 years or older. For those with CD4+ cell counts greater than 200/µL and less than or equal to 350/µL, for whom the guidelines say “Treat; deferring may be appropriate” regardless of HIV RNA level, the simulation would advise to treat, deferring only if the patient is 40 years or older and has an HIV RNA level less than 100,000 copies/mL or if the patient is 50 years or older and has an HIV RNA level over 100,000 copies/mL.

Given the complexity and volume of new information relevant to HIV care and patient survival, such as that generated above, it might seem self-evident that HIV specialists are critical to implementation. Indeed, a dozen years ago, a study confirmed the impression that physician experience was the major factor in extending the life expectancy of an HIV-positive person.6 In an age that increasingly relies on hospitalists, a recent study modifies that premise.7 Provider type or attending physician experience with HIV-infected inpatients, reviewed at 6 academic institutions in geographically diverse sites from 2001 to 2003, had minimal effect on quality of care.7 The use of multidisciplinary inpatient teams may be more pertinent.

Records from 1207 HIV-positive patients cared for by 214 physicians, 43 of whom were hospitalists, were reviewed. The 3 most common discharge diagnoses were pneumonia, asthma, and pancreatitis. Physician experience was not associated with any outcome measure, including coordination with primary care physician, pain control, pneumococcal vaccination, and length of stay.7

The success of highly active antiretroviral therapy, at least where readily available, at no cost, and with clinical laboratory support, has also diminished concerns about the spread of drug-resistant strains. For example, in British Columbia, antiretroviral therapy and resistance testing are provided free of charge. A recent review of 7363 patients treated there from 1996 to 2007 found a dramatic decrease in the incidence of new drug resistance against PIs; NNRTIs; and NRTIs, including lamivudine-from 7% to 15% in 2000 to 0.8% to 2% in 2007.8 More than 90% of treated patients had either suppressed plasma HIV RNA to undetectable levels or had drug-sensitive HIV in their most recent test results.8 The tragedy for this population, however, was the simultaneous announcement that 39% of the 1436 persons who died of AIDS-related conditions in British Columbia from 1997 to 2005 never received highly active antiretroviral therapy.9

As in the United States, about 25% of HIV-infected persons in Canada are still unaware they are HIV-infected.10 Computer modeling to optimize treatment, together with continued development of new classes of anti-HIV medications, is wonderful. But improved outreach and better prevention methods are also clearly needed.


References1. Chun TW, Justement JS, Moir S, et al. Decay of the HIV reservoir in patients receiving antiretroviral therapy for extended periods: implications for eradication of virus. J Infect Dis. 2007;195:1762-1764.
2. Braithwaite RS, Roberts MS, Chang CC, et al. Influence of alternative thresholds for initiating HIV treatment on quality-adjusted life expectancy: a decision model. Ann Intern Med. 2008;148:178-185.
3. Panel on Clinical Practices for the Treatment of HIV Infection. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services. January 29, 2008;1-128.
4. Madden E, Lee G, Kotler DP, et al. Association of antiretroviral therapy with fibrinogen levels in HIV-infection. AIDS. 2008;22:707-715.
5. Braithwaite RS, Justice AC, Chang CC, et al. Estimating the proportion of patients infected with HIV who will die of comorbid diseases. Am J Med. 2005;118: 890-898.
6. Kitahata MM, Koepsell TD, Deyo RA, et al. Physicians' experience with the acquired immunodeficiency syndrome as a factor in patients' survival. N Engl J Med. 1996;334:701-706.
7. Schneider JA, Zhang Q, Auerbach A, et al. Do hospitalists or physicians with greater inpatient HIV experience improve HIV care in the era of highly active antiretroviral therapy? Results from a multicenter trial of academic hospitalists. Clin Infect Dis. 2008;46:1085-1092.
8. Lima V, Hudson E, Wynhoven B, et al. Drastically declining incidence of HIV drug resistance: the end of the beginning? 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston. Abstract 895.
9. Lazaruk S. Hundreds dying of AIDS never accessed drug cocktails: report. Vancouver Province. February 23, 2008. news/story.html?id=82584aa1-06c7-4e4d-9d76-8222901d041a&k=39033. Accessed May 14, 2008.
10. Moore D. 40% of HIV sufferers died without getting help, study finds. Globe and Mail. February 23, 2008:S2.